Replication methods for variance estimation draw multiple replicates (subsamples) from the full sample by following a specific resampling scheme. Commonly used resampling schemes include bootstrap, balanced repeated replication (BRR), and jackknife. PROC SURVEYFREQ estimates the parameter of interest (a proportion, total, odds ratio, or other statistic) from each replicate, and then uses the variability among replicate estimates to estimate the overall variance of the parameter estimate. For more information, see Wolter (1985) and Lohr (2010).
The balanced repeated replication (BRR) variance estimation method requires a stratified sample design with two PSUs in each stratum. Each replicate is created by deleting one PSU per stratum according to the corresponding Hadamard matrix and adjusting the original weights of the remaining PSUs. The adjusted weights are called replicate weights. PROC SURVEYFREQ also provides Fay’s BRR method, which is a modification of the BRR method. For more information, see the sections Balanced Repeated Replication (BRR) Method and Fay’s BRR Method.
The jackknife method deletes one PSU at a time from the full sample to create replicates and adjusts the original weights to obtain replicate weights. The total number of replicates is the number of PSUs. If the sample design is stratified, each stratum must contain at least two PSUs, and the jackknife method is applied separately within each stratum. For more information, see the section Jackknife Method.
Instead of having PROC SURVEYFREQ construct replicate weights for the analysis, you can input your own replicate weights by using a REPWEIGHTS statement. This can be useful if you need to do multiple analyses with the same set of replicate weights, or if you have access to replicate weights without complete design information. For more information, see the REPWEIGHTS statement.